TRAFFIC_LOGSTATS(8)

✨ TRAFFIC_LOGSTATS ✨

ok so bestie, welcome to traffic_logstats — the Traffic Server analyzer from the Apache Traffic Server manual (section 8, we're in the big-kid sysadmin pages 💅). Pull up a chair, grab a strawberry 🍓, and let's get into it.

---

🌸 SYNOPSIS 🌸

traffic_logstats [options]

that's it. that's the whole vibe. options go in, stats come out. ✨

---

💖 DESCRIPTION 💖

so traffic_logstats is a log parsing utility, and its whole ~thing~ is producing metrics for total requests and per-origin requests. 🎀

ngl, right now it only supports parsing and processing the Squid binary log format, OR a custom format that's compatible with the initial log fields of the Squid format. so like, don't show up with some random format and expect it to just work, bestie. 🙈

the output? you've got options, queen:

• a human readable text file 📄
• or JSON format 🧚

and here's the iconic part — parsing can be done incrementally, and traffic_logstats supports restarting where it left off previously (the state is stored in an external file). so it's giving main character energy with actual memory. periodt. 💅 this is super useful when you're collecting metrics periodically into a stats processing system, and it also handles the case where a log file gets rotated. no drama, it just picks back up. 🦋

⚠️ okay important heads-up: the per-URL metrics (-u) require that you specify a size of the LRU used for keeping the counters. this is so traffic_logstats doesn't consume an exorbitant amount of memory. don't skip this one, bestie, your RAM will simply not slay. 💀

---

👑 OPTIONS 👑

• -f FILE, --log_file FILE — the specific logfile to parse 📂
• -o LIST, --origin_list LIST — only show stats for listed Origins
• -O FILE, --origin_file FILE — file listing Origins to show
• -M COUNT, --max_origins COUNT — max number of Origins to show
• -u COUNT, --urls COUNT — produce JSON stats for URLs; the argument is the LRU size 🧠
• -U COUNT, --show_urls COUNT — only show max this number of URLs
• -A, --as_object — produce URL stats as a JSON object instead of an array 🎁
• -C, --concise — eliminate metrics that can be inferred from other values ✂️
• -i, --incremental — incremental log parsing 🔁
• -S FILE, --statetag FILE — name of the state file to use 💾
• -t, --tail — parse the last <sec> seconds of log ⏱️
• -s, --summary — only produce the summary 📝
• -j, --json — produce JSON formatted output 🧚
• -c, --cgi — produce HTTP headers suitable as a CGI 🌐
• -m, --min_hits — minimum total hits for an Origin
• -a, --max_age — max age for log entries to be considered ⏳
• -l COUNT, --line_len COUNT — output line length 📏
• -T TAGS, --debug_tags TAGS — colon-separated Debug Tags 🏷️
• -r, --report_per_user — report stats per username of the authenticated client caun instead of host, see squid log format 👤
• -n, --no_format_check — don't validate the log format field names according to the squid log format. this would allow squid format fields to be replaced, i.e. the username of the authenticated client caun with a random header value by using cqh, or to remove the client's host IP address from the log for privacy reasons. 🙈
• -h, --help — print usage information and exit. 🫶
• -V, --version — print version information and exit. ⭐

---

🎀 COPYRIGHT 🎀

2026, dev@trafficserver.apache.org

---

and that's the whole page, bestie!! 🦋✨ this was version 9.2, dated Feb 12, 2026, brought to you by the Apache Traffic Server girlies — go parse some logs and slay. 💖👑